Merge branch 'master'; commit 'v2.6.38-rc7' into next

7 files changed, 59 insertions, 130 deletions

diff --git a/kernel/irq/Kconfig b/kernel/irq/Kconfig
index 31d766bf5d2e..8e42fec7686d 100644
--- a/kernel/irq/Kconfig
+++ b/kernel/irq/Kconfig
@@ -9,9 +9,6 @@ menu "IRQ subsystem"
 config GENERIC_HARDIRQS
        def_bool y
 
-config GENERIC_HARDIRQS_NO__DO_IRQ
-       def_bool y
-
 # Select this to disable the deprecated stuff
 config GENERIC_HARDIRQS_NO_DEPRECATED
        def_bool n
diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c
index e2347eb63306..3540a7190122 100644
--- a/kernel/irq/handle.c
+++ b/kernel/irq/handle.c
@@ -118,114 +118,3 @@ irqreturn_t handle_IRQ_event(unsigned int irq, struct irqaction *action)
 
 	return retval;
 }
-
-#ifndef CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ
-
-#ifdef CONFIG_ENABLE_WARN_DEPRECATED
-# warning __do_IRQ is deprecated. Please convert to proper flow handlers
-#endif
-
-/**
- * __do_IRQ - original all in one highlevel IRQ handler
- * @irq:	the interrupt number
- *
- * __do_IRQ handles all normal device IRQ's (the special
- * SMP cross-CPU interrupts have their own specific
- * handlers).
- *
- * This is the original x86 implementation which is used for every
- * interrupt type.
- */
-unsigned int __do_IRQ(unsigned int irq)
-{
-	struct irq_desc *desc = irq_to_desc(irq);
-	struct irqaction *action;
-	unsigned int status;
-
-	kstat_incr_irqs_this_cpu(irq, desc);
-
-	if (CHECK_IRQ_PER_CPU(desc->status)) {
-		irqreturn_t action_ret;
-
-		/*
-		 * No locking required for CPU-local interrupts:
-		 */
-		if (desc->irq_data.chip->ack)
-			desc->irq_data.chip->ack(irq);
-		if (likely(!(desc->status & IRQ_DISABLED))) {
-			action_ret = handle_IRQ_event(irq, desc->action);
-			if (!noirqdebug)
-				note_interrupt(irq, desc, action_ret);
-		}
-		desc->irq_data.chip->end(irq);
-		return 1;
-	}
-
-	raw_spin_lock(&desc->lock);
-	if (desc->irq_data.chip->ack)
-		desc->irq_data.chip->ack(irq);
-	/*
-	 * REPLAY is when Linux resends an IRQ that was dropped earlier
-	 * WAITING is used by probe to mark irqs that are being tested
-	 */
-	status = desc->status & ~(IRQ_REPLAY | IRQ_WAITING);
-	status |= IRQ_PENDING; /* we _want_ to handle it */
-
-	/*
-	 * If the IRQ is disabled for whatever reason, we cannot
-	 * use the action we have.
-	 */
-	action = NULL;
-	if (likely(!(status & (IRQ_DISABLED | IRQ_INPROGRESS)))) {
-		action = desc->action;
-		status &= ~IRQ_PENDING; /* we commit to handling */
-		status |= IRQ_INPROGRESS; /* we are handling it */
-	}
-	desc->status = status;
-
-	/*
-	 * If there is no IRQ handler or it was disabled, exit early.
-	 * Since we set PENDING, if another processor is handling
-	 * a different instance of this same irq, the other processor
-	 * will take care of it.
-	 */
-	if (unlikely(!action))
-		goto out;
-
-	/*
-	 * Edge triggered interrupts need to remember
-	 * pending events.
-	 * This applies to any hw interrupts that allow a second
-	 * instance of the same irq to arrive while we are in do_IRQ
-	 * or in the handler. But the code here only handles the _second_
-	 * instance of the irq, not the third or fourth. So it is mostly
-	 * useful for irq hardware that does not mask cleanly in an
-	 * SMP environment.
-	 */
-	for (;;) {
-		irqreturn_t action_ret;
-
-		raw_spin_unlock(&desc->lock);
-
-		action_ret = handle_IRQ_event(irq, action);
-		if (!noirqdebug)
-			note_interrupt(irq, desc, action_ret);
-
-		raw_spin_lock(&desc->lock);
-		if (likely(!(desc->status & IRQ_PENDING)))
-			break;
-		desc->status &= ~IRQ_PENDING;
-	}
-	desc->status &= ~IRQ_INPROGRESS;
-
-out:
-	/*
-	 * The ->end() handler has to deal with interrupts which got
-	 * disabled while the handler was running.
-	 */
-	desc->irq_data.chip->end(irq);
-	raw_spin_unlock(&desc->lock);
-
-	return 1;
-}
-#endif
diff --git a/kernel/irq/internals.h b/kernel/irq/internals.h
index 4571ae7e085a..99c3bc8a6fb4 100644
--- a/kernel/irq/internals.h
+++ b/kernel/irq/internals.h
@@ -3,6 +3,12 @@
  */
 #include <linux/irqdesc.h>
 
+#ifdef CONFIG_SPARSE_IRQ
+# define IRQ_BITMAP_BITS	(NR_IRQS + 8196)
+#else
+# define IRQ_BITMAP_BITS	NR_IRQS
+#endif
+
 extern int noirqdebug;
 
 #define irq_data_to_desc(data)	container_of(data, struct irq_desc, irq_data)
diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c
index 9988d03797f5..2039bea31bdf 100644
--- a/kernel/irq/irqdesc.c
+++ b/kernel/irq/irqdesc.c
@@ -72,6 +72,8 @@ static inline int desc_node(struct irq_desc *desc) { return 0; }
 
 static void desc_set_defaults(unsigned int irq, struct irq_desc *desc, int node)
 {
+	int cpu;
+
 	desc->irq_data.irq = irq;
 	desc->irq_data.chip = &no_irq_chip;
 	desc->irq_data.chip_data = NULL;
@@ -83,7 +85,8 @@ static void desc_set_defaults(unsigned int irq, struct irq_desc *desc, int node)
 	desc->irq_count = 0;
 	desc->irqs_unhandled = 0;
 	desc->name = NULL;
-	memset(desc->kstat_irqs, 0, nr_cpu_ids * sizeof(*(desc->kstat_irqs)));
+	for_each_possible_cpu(cpu)
+		*per_cpu_ptr(desc->kstat_irqs, cpu) = 0;
 	desc_smp_init(desc, node);
 }
 
@@ -91,7 +94,7 @@ int nr_irqs = NR_IRQS;
 EXPORT_SYMBOL_GPL(nr_irqs);
 
 static DEFINE_MUTEX(sparse_irq_lock);
-static DECLARE_BITMAP(allocated_irqs, NR_IRQS);
+static DECLARE_BITMAP(allocated_irqs, IRQ_BITMAP_BITS);
 
 #ifdef CONFIG_SPARSE_IRQ
 
@@ -133,8 +136,7 @@ static struct irq_desc *alloc_desc(int irq, int node)
 	if (!desc)
 		return NULL;
 	/* allocate based on nr_cpu_ids */
-	desc->kstat_irqs = kzalloc_node(nr_cpu_ids * sizeof(*desc->kstat_irqs),
-					 gfp, node);
+	desc->kstat_irqs = alloc_percpu(unsigned int);
 	if (!desc->kstat_irqs)
 		goto err_desc;
 
@@ -149,7 +151,7 @@ static struct irq_desc *alloc_desc(int irq, int node)
 	return desc;
 
 err_kstat:
-	kfree(desc->kstat_irqs);
+	free_percpu(desc->kstat_irqs);
 err_desc:
 	kfree(desc);
 	return NULL;
@@ -166,7 +168,7 @@ static void free_desc(unsigned int irq)
 	mutex_unlock(&sparse_irq_lock);
 
 	free_masks(desc);
-	kfree(desc->kstat_irqs);
+	free_percpu(desc->kstat_irqs);
 	kfree(desc);
 }
 
@@ -215,6 +217,15 @@ int __init early_irq_init(void)
 	initcnt = arch_probe_nr_irqs();
 	printk(KERN_INFO "NR_IRQS:%d nr_irqs:%d %d\n", NR_IRQS, nr_irqs, initcnt);
 
+	if (WARN_ON(nr_irqs > IRQ_BITMAP_BITS))
+		nr_irqs = IRQ_BITMAP_BITS;
+
+	if (WARN_ON(initcnt > IRQ_BITMAP_BITS))
+		initcnt = IRQ_BITMAP_BITS;
+
+	if (initcnt > nr_irqs)
+		nr_irqs = initcnt;
+
 	for (i = 0; i < initcnt; i++) {
 		desc = alloc_desc(i, node);
 		set_bit(i, allocated_irqs);
@@ -234,7 +245,6 @@ struct irq_desc irq_desc[NR_IRQS] __cacheline_aligned_in_smp = {
 	}
 };
 
-static unsigned int kstat_irqs_all[NR_IRQS][NR_CPUS];
 int __init early_irq_init(void)
 {
 	int count, i, node = first_online_node;
@@ -250,7 +260,8 @@ int __init early_irq_init(void)
 	for (i = 0; i < count; i++) {
 		desc[i].irq_data.irq = i;
 		desc[i].irq_data.chip = &no_irq_chip;
-		desc[i].kstat_irqs = kstat_irqs_all[i];
+		/* TODO : do this allocation on-demand ... */
+		desc[i].kstat_irqs = alloc_percpu(unsigned int);
 		alloc_masks(desc + i, GFP_KERNEL, node);
 		desc_smp_init(desc + i, node);
 		lockdep_set_class(&desc[i].lock, &irq_desc_lock_class);
@@ -275,6 +286,22 @@ static void free_desc(unsigned int irq)
 
 static inline int alloc_descs(unsigned int start, unsigned int cnt, int node)
 {
+#if defined(CONFIG_KSTAT_IRQS_ONDEMAND)
+	struct irq_desc *desc;
+	unsigned int i;
+
+	for (i = 0; i < cnt; i++) {
+		desc = irq_to_desc(start + i);
+		if (desc && !desc->kstat_irqs) {
+			unsigned int __percpu *stats = alloc_percpu(unsigned int);
+
+			if (!stats)
+				return -1;
+			if (cmpxchg(&desc->kstat_irqs, NULL, stats) != NULL)
+				free_percpu(stats);
+		}
+	}
+#endif
 	return start;
 }
 #endif /* !CONFIG_SPARSE_IRQ */
@@ -391,7 +418,9 @@ void dynamic_irq_cleanup(unsigned int irq)
 unsigned int kstat_irqs_cpu(unsigned int irq, int cpu)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
-	return desc ? desc->kstat_irqs[cpu] : 0;
+
+	return desc && desc->kstat_irqs ?
+			*per_cpu_ptr(desc->kstat_irqs, cpu) : 0;
 }
 
 #ifdef CONFIG_GENERIC_HARDIRQS
@@ -401,10 +430,10 @@ unsigned int kstat_irqs(unsigned int irq)
 	int cpu;
 	int sum = 0;
 
-	if (!desc)
+	if (!desc || !desc->kstat_irqs)
 		return 0;
 	for_each_possible_cpu(cpu)
-		sum += desc->kstat_irqs[cpu];
+		sum += *per_cpu_ptr(desc->kstat_irqs, cpu);
 	return sum;
 }
 #endif /* CONFIG_GENERIC_HARDIRQS */
diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
index 0caa59f747dd..9033c1c70828 100644
--- a/kernel/irq/manage.c
+++ b/kernel/irq/manage.c
@@ -1100,7 +1100,7 @@ int request_threaded_irq(unsigned int irq, irq_handler_t handler,
 	if (retval)
 		kfree(action);
 
-#ifdef CONFIG_DEBUG_SHIRQ
+#ifdef CONFIG_DEBUG_SHIRQ_FIXME
 	if (!retval && (irqflags & IRQF_SHARED)) {
 		/*
 		 * It's a shared IRQ -- the driver ought to be prepared for it
diff --git a/kernel/irq/migration.c b/kernel/irq/migration.c
index 1d2541940480..441fd629ff04 100644
--- a/kernel/irq/migration.c
+++ b/kernel/irq/migration.c
@@ -56,6 +56,7 @@ void move_masked_irq(int irq)
 void move_native_irq(int irq)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
+	bool masked;
 
 	if (likely(!(desc->status & IRQ_MOVE_PENDING)))
 		return;
@@ -63,8 +64,15 @@ void move_native_irq(int irq)
 	if (unlikely(desc->status & IRQ_DISABLED))
 		return;
 
-	desc->irq_data.chip->irq_mask(&desc->irq_data);
+	/*
+	 * Be careful vs. already masked interrupts. If this is a
+	 * threaded interrupt with ONESHOT set, we can end up with an
+	 * interrupt storm.
+	 */
+	masked = desc->status & IRQ_MASKED;
+	if (!masked)
+		desc->irq_data.chip->irq_mask(&desc->irq_data);
 	move_masked_irq(irq);
-	desc->irq_data.chip->irq_unmask(&desc->irq_data);
+	if (!masked)
+		desc->irq_data.chip->irq_unmask(&desc->irq_data);
 }
-
diff --git a/kernel/irq/resend.c b/kernel/irq/resend.c
index 891115a929aa..dc49358b73fa 100644
--- a/kernel/irq/resend.c
+++ b/kernel/irq/resend.c
@@ -23,7 +23,7 @@
 #ifdef CONFIG_HARDIRQS_SW_RESEND
 
 /* Bitmap to handle software resend of interrupts: */
-static DECLARE_BITMAP(irqs_resend, NR_IRQS);
+static DECLARE_BITMAP(irqs_resend, IRQ_BITMAP_BITS);
 
 /*
  * Run software resends of IRQ's